Quality of life, Survival, COmorbidities, and Pharmacoeconomics in Elderly patients with Acute Myeloid Leukemia

Acute myeloid leukemia (AML) is a malignant disease of the bone marrow with a
poor prognosis, especially in elderly patients [1][2]. With a year incidence of
3 - 4,5 cases per 100.000 people AML is a rare disease [3]. As the median age
at diagnosis is 68 years, AML at a young age is viewed as a different disease
entity [1][2][3]. Patients aged below 65 are commonly treated with intensive
induction chemotherapy until remission or remission with minimal residual
disease is reached. Remission, if reached, is then consolidated with an
allogenic or autologic stem cell transplant. Elderly patients are rarely in a
good enough baseline health state to undergo this highly intensive, somewhat
dangerous, and overall expensive treatment trajectory [1][3]. In 2012 and 2015
respectively the hypomethylating agents Azacitidine AZA and Decitabine DEC were
introduced to the market and offer less intensive treatment options for these
elderly patients [4][5]. Because of this, stratifying de novo AML patients into
the following three groups by the so called Ferrara-criteria became common
practice: Fit for intensive chemotherapy, not fit for intensive chemotherapy
but fit enough for HMA based therapy, and unfit even for HMA based therapy [6].
After a period in which HMA monotherapy was the first choice of treatment for
patients deemed ineligible for intensive induction therapy, the combination of
a HMA with Venetoclax VEN was proven to lead to superior outcomes and has been
promoted to first choice of treatment for this patient group in March 2022 [7].
Combination of a HMA with Ivosidenib in patients with a specific cytogenetic
lesion has proven even superior to HMA+VEN, but remains a 2nd line of treatment
for now [8]. In the near future, a multitude of other agents, currently in
phase III clinical trials, are expected to make it into treatment guidelines
[8][9][10]. All of these agents have in common that they are researched as
additions to either an HMA or even HMA+VEN backbone. These HMA based
combination regimens are expected to become the main weapons in the therapeutic
arsenal for AML as they seem to improve survival when compared against HMA
monotherapy. In the clinical trials investigating these regimens, QOL
questionnaires such as the EORTC-QLQ-C30 are commonly collected. Results of
this data are however often left unpublished or are only published years later
and in lower impact factor journals. Real world experience in the MCL with for
example the AZA+VEN regimen points towards a noticeable increase in treatment
intensity as patients spend more time in the hospital, undergo more bonemarrow
analysis, etc. Data from phase III clinical trials point towards noticeable
increases in treatment toxicity (side effects & adverse events)
[7][8][9][10][11]. Overall a negative effect on QOL seems plausible. Data of a
patients* health state before and at diagnosis can be predictive of outcomes
such as survival and quality of life. Such data may include patient related
factors such as performance status and comorbidity profile, as well as disease
related factors such as mutation profile and bloodvalues. Multiple attempts
have been made in the past to construct prognostic tools for various groups of
AML patients (Wheatley Index, HCT-CI, AML-CM, ALFA1200, Liu-Index)
[6][12][13][14][15][16][17][18]. Being able to predict which patients do or do
not develop serious adverse events or decreases in quality of life may aid in
identifying patients that may benefit more or may benefit less from HMA
combination regimens. The first HMA combination regimen that achieved approval
for reimbursement for treatment of AML in the Netherlands is HMA+VEN. The
evaluation of novel agents or regiments through the Zorginstituut Nederland ZIN
include a detailed pharmacoeconomic model, attempting to quantify both benefits
as well as costs. In the case of HMA+VEN, quality of life was calculated based
upon the proportional shortfalls method, estimating that the average AML
patient loses 8,39 QALYs through the disease [19]. Furthermore they estimate
that HMA+VEN will result in a gain of 1,72 QALYs compared to the 0,75 QALYs
gained by HMA monotherapy. The data for the HMA+VEN intervention for this model
comes solely from patent holder AbbVie. Furthermore, the commencing cost
effectiveness calculations are lacking in respect to quantification of time
spend in the hospital, and transfusion burden. Therefore it is deemed of high
interest to recalculate the QALY data based upon real-world survival and
quality of life data, as well as to recalculate the ICER of the HMA+VEN regimen
against HMA monotherapy while including prospectively measured additional costs
in the form of healthcare consumption (such as transfusion usage and length of
hospital stay). For this purpose we aim to measure baseline characteristics,
quality of life, healthcare consumption, and survival prospectively in an
unselected real-world population of elderly AML patients treated with HMA
combination regimens.

We aim to measure quality of life and overall survival in patients with AML
classified as ineligible for intensive chemotherapy and get treated with HMA
combination regimens in the Medical Centre Leeuwarden. We also aim to determine
the predictive potential of various comorbidity indexes and clinical tools for
assessing a patients* health state on outcomes such as their survival.
Furthermore we aim to gather data on healthcare consumption so that insights
into survival, quality of life, and costs can be combined into a
pharmacoeconomic analysis of the novel regimens.

The QSCOPE-AML study is a prospective, observational, single-centre, real world
study. All de novo AML patients aged 65 or older that are classified as *not
eligible for intensive chemotherapy* by the list 1 of the Ferrara criteria or
by assessment of their haematologist and are treated with a HMA combination
regimen in the Medical Centre Leeuwarden are included in the study. Patients
are approached for participation in the study by the researchers with a patient
information leaflet PIL and an informed consent form is signed. Patients are
followed prospectively until either loss-to-follow-up or death. The EQ-5D-5L,
EORTC-QLQ-C30 and FACIT questionnaires are taken before start of chemotherapy
and again after 2, 4, 8, 12, 18, and 24 months. The patients priority
questionnaire is taken once before start of chemotherapy. For assessing
baseline characteristics and comorbidity scores the patients* medical history
the EHR/EPD is reviewed. Time spend in the hospital (planned, unplanned,
consultations, nights) is measured per cycle of chemotherapy starting from
diagnosis. Transfusion dependence is measured cumulatively over the intervals
0-2, 2-4, 4-8, 8-12, 12-18, and 18-24 months.

Patients are approached with the following four questionnaires:
• The general 5 item EQ-5D-5L
• The fatigue focussed 7 item PROMIS-7SFa
• The cancerspecific 30 item EORTC-QLQ-C30
• The question to order 4 possible treatment goals by importance

After inclusion, patients are approached with all 4 questionnaires. For all
items of the EQ-5D, PROMIS, and EORTC questionnares patients are asked to rate
their life/health now and 1 month prior to diagnosis. For this initial
measurement 45 minutes are deemed realistic.

2, 4, 8, 12, 18, and 24 month after diagnosis patients are approached again
with the EQ-5D, PROMIS, and EORTC questionnaires. All 42 questions (5+7+30) are
of shot and definitive character and have between 4 and 7 (commonly 5)
answering options. The researchers deem 30 minutes per measurement per patient
a realistic estimate of timeburden.